• Volume/Page
  • Keyword
  • DOI
  • Citation
  • Advanced
   
 
 
 

Journal of the Acoustical Society of America

SECTION LISTING

BROWSE VOLUMES

Year Range: 
Search Issue | RSS Feeds RSS
Previous Issue Next Issue

Nov 1976

Volume 60, Issue S1, pp. S1-S125

Return to All Sections
back to top
RSS Feeds
back to top Session VV. Psychological and Physiological Acoustics VIII: Pitch and Detection
Contributed Papers
FREE

Monaural detection of complex transient “real world” sounds (A)

G. Richard Price and David C. Hodge

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S116-S116 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Given a long‐term interest in evaluating auditory performance in the Army context, a first step was taken by examining the ability of individual ears to detect complex transient sounds, such as those produced by personnel movement or items of personnel combat equipment. A Fournier analysis was performed on each of 24 sounds every 20 msec and the energy was integrated into critical bands. In 20‐msec steps within each critical band, the energy was integrated for 200‐ and the 200‐msec period with the most energy in it was selected as the one most likely to be detected. These spectra were compared graphically with the auditory thresholds of 20 ears (selected to have a wide range of hearing levels) measured at critical‐band center frequencies. The relative level at which the energy is one critical band equalled or exceeded the measured threshold for that critical band was selected and the detection levels were measured by having subjects track their thresholds for the complex sound. The correlations between the predicted and actual detection levels ranged from 0.89 to 0.98 with a 0.94 mean. Because of the generally broad distribution of energy in the transient sounds and the sensitivity of the ears in the midrange, most of the detections were a function of energy in this region. Ears with large high‐frequency losses therefore did not perform much worse than the normal ears. If background noise typical of the world were included in the predictive procedure, then differences between ears would have been smaller.
FREE

Signal detection in time varying noise backgrounds (A)

Sanford Fidell

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S116-S116 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Prediction of the detectability of well‐defined signals in steady‐state background noises are based on considerable theory and much practical experience. Little information is presently available to support predictions of detectability of complex signals in time‐varying noise environments. The current study has temporal variability of background noise is long with respect to signal duration, observers' detection performance improves in more highly variable noise environments of constant average energy. Observers detecting a sinusoid embedded in one‐third‐octave band of Gaussian noise in a two alternative forced‐choice task produced psychometric functions characterized by higher d′ values for increasingly variable background noises, relative to performance in steady‐state noise.
FREE

Effects of forward masking on detection and intensity discrimination of pulsed sinusoids (A)

Gregory P. Wildin and Neal F. Viemeister

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S116-S116 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
The effect of the intensity of a pure‐tone masker on the forward‐masked threshold of a brief pure‐tone signal of the same frequency was investigated using masker level up to 100 dB SPL. Results agree with and extend those of Houtgast [in Facts and Models in Hearing, edited by E. Zwicker and E. Terhardt (Springer, Berlin, 1974), pp. 258–265 and are described by a square‐root relation between masker and signal level, to masker levels of at least 80 dB. In contrast, Weber's law is approximately valid for intensity discrimination of pulsed sinusoids in the presence of a pure‐tone forward masker when the ratio of masker level of the standard is fixed. For fixed intensities above the detection threshold in forward masking, the forward‐masked increment threshold generally increases only slightly with increases in masker level. [Research supported by the Center for Research in Human Learning and by NSF.]
FREE

Estimating the asymmetry of the auditory filter (A)

Roy D. Patterson and Ian Nimmo‐Smith

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S116-S116 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Although recent experiments have not revealed any marked asymmetry in the shape of the filter, some derivation assume a symmetric filter. To determine the degree of asymmetry more precisely, tone threshold was repeatedly measured as a notch in a wide‐band noise was moved past the tone. The width of the notch was also varied. The asymmetry of the resulting threshold curves revealed a small asymmetry in the filter shape. In attempting to generalize earlier methods of estimating filter shape to the present data, it was discovered that the analytic statement of the problem is in a class mathematicians have labelled “ill‐posed” because the “solution” filter is extremely sensitive to small perturbations of the threshold data. To improve the formulation of the problem a new and hopefully more useful class of filter shapes has been adopted.
FREE

Effect of stimulus variability on auditory filter shape (A)

Roy D. Patterson and G. Bruce Henning

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S116-S116 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
In recent derivations of auditory filter shape based on tone‐in‐noise detection experiments, the noise is represented by its power spectrum; that is, it has been assumed that filter shape is largely independent of masker variance. In an attempt to extend the generality of the filter‐shape concept, one of the experiments has been replicated using a tonal maker in hopes of demonstrating that filter shapes derived with noise maskers are applicable when the masker variability is markedly reduced. Although some form of detection mechanism must follow the auditory filter, previous studies have not considered the mechanism explicitly. We will discuss the implications of including an energy detector at the output of the filter for the two‐tone and noise‐masking experiments, and in so doing, will outline the conditions under which changes in stimulus variability will affect performance and through it, the derived filter shapes.
FREE

On the growth of the pulsation threshold of a suppressed tone (A)

Robert V. Shannon

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S117-S117 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Consider a 1200‐Hz tone suppressing a 1000‐Hz tone. For a fixed level of the 1200‐Hz tone (50 dB) the pulsation threshold was measured at 1000 Hz as the 1000‐Hz tone in the “masker” was raised from absolute threshold to 60 dB in 5‐dB steps. When the level of the 1000‐Hz tone was less than the simultaneous masked threshold at 1000 Hz the pulsation threshold measured was the same as if the 1000‐Hz tone were not present in the “masker”. When the level of the 1000‐Hz tone was raised above the simultaneous masked‐threshold level the pulsation‐threshold level increased at the same rate as the physical level—maintaining a constant number of decibels of suppression. When the level of the 1000‐Hz tone was raised above the level of the 1200‐Hz tone the pulsation‐threshold level increased faster than the physical level, showing “recruitment” of pulsation threshold or a reduction in suppression. [Work supported by NINCDS.]
FREE

Suppression in pulsation‐threshold patterns (A)

Robert V. Shannon

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S117-S117 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
In a pulsation, threshold experiment the frequencies and intensities of the two tones in the “masker” were fixed and the frequency of the signal was varied. The plot of pulsation threshold versus signal frequency is called the pulsation‐threshold pattern of the “masker”. First the pulsation‐threshold pattern was measured when the “masker” contained only a 1000‐Hz tone. Then the patterns were measured for a 400‐Hz tone and for a 1500‐Hz tone. Finally, the pattern was measured when the masker contained both the 1000‐Hz tone and either the 400‐Hz tone or the 1500‐Hz tone. The 400‐Hz tone was able to reduce the pulsation thresholds around 1000‐Hz by 20 dB or more for each of the three subjects. The 1500‐Hz tone did not reduce the pulsation thresholds at 1000 Hz but did at 1100 Hz, i.e., on the high‐frequency tail of the 1000‐Hz pattern. Suppression is assumed to be the cause of these reductions. Thus one tone is able to suppress only part of the excitation pattern of another tone. Mutual suppression was also indicated for one subject. Relationships to partial masking will be discussed. [Work supported the NINCDS.]
FREE

Measurement of a binaural temporal transfer function (A)

D. W. Grantham and F. L. Wightman

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S117-S117 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
In order to derive a kind of “binaural temporal transfer function” for the auditory system, observers detected a stable target (a binaural click presented with an interaural time difference of +500 μsec) imbedded in a “moving” masker (a noise presented with an interaural delay sinusoidally varied from +500 to −500 μsec). Detectability of the click was measured as a function of the “position” of the noise at the time the click was presented and the rate at which the noise “moved”. As was expected, when the noise moved very slowly, less than 1 Hz, or not at all, detectability was best when the click appeared at one ear and the noise appeared at the other, and poorest when they both appeared at the same ear. At high rates (5 Hz) detectability was independent of the relative positions of the click and the noise. [This work was supported in part by NIH Grant NS12045.]
FREE

Modulation thresholds and temporal modulation transfer functions (A)

Neal F. Viemeister

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S117-S117 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Temporal Modulation Transfer Functions (TMTF's) were obtained by measuring the threshold amplitude of sinusoidal modulation as a function of modulating frequency. For modulation frequencies below approximately 800 Hz, TMTF's obtained with a continuous wide‐band noise carrier generally show the low‐pass characteristic reported previously [J. Acoust. Soc. Am. 53, 314(A) (1973)], That is, with increasing modulation frequency the amplitude of modulation required for threshold remains constant up to approximately 10 Hz and then increases monotonically up to 800 Hz. The interpretation is that at high modulation frequencies the auditory system temporally “smooths” the amplitude fluctuations produced by modulation and the observer therefore requires greater modulation amplitude at the input in order to detect the modulation. For modulation frequencies greater than 800 Hz, modulation threshold is independent of modulation frequency and can be predicted from the increment threshold for wide‐band noise. The form of the empirical TMTF generally agrees with that predicted by the familiar model consisting of half‐wave rectification followed by “leaky integration.” The time constant of the integrator is estimated to be 3 msec. [Research supported by NIH.]
FREE

Structural rules for the perception of the periodicity of repeated random waveforms (A)

Irwin Pollack

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S117-S117 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Two structural rules underlying the periodicity of repeated random waveforms were examined: the number of successive copies of the to‐be‐repeated elements; and phase inversion of the to‐be‐repeated elements. Two behavioral measures of periodicity strength were defined: the accuracy of pitch matching and the fraction of independently varying elements which must be inserted with a repeated random pattern for a threshold level of discrimination from a nonrandom pattern. A variety of autocorrelation measures were performed upon the signals. The signal measure most closely related to behavioral measures of periodicity strength is the maximum autocorrelation among the first 20 periodic delay intervals. Additional weightings with the average autocorrelation across the 20 delay intervals failed to improve further the relation with periodicity strength.
FREE

Pitch strength of ripple noise (A)

William A. Yost and Richard Hill

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S117-S117 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Ripple noise is produced when a broadband noise is delayed (T) and either added to or subtracted from the original noise. The resulting waveform has a power spectrum in which power varies cosinusoidally as a function of frequency. Ripple noise has been used to study pitch perception and echo processing. In the present experiments subjects were asked to discriminate between two noises which differed in the amount of delay introduced (one stimulus continued a delay T the second stimulus a delay T + 0.09T). The discriminability of these two noises was studied as a function of the amount of attenuation added to the delayed noise. Discriminations between these two noises could be made when T ranged from 0.5 to 20 msec before the subjects could no longer make the discrimin‐delayed noise when the delays were in the range from 2 to 5 msec before the subjects could no longer make the discriminations. More attenuation could be added to the delayed noise for threshold discrimination when it was added back to the original source than when it was subtracted from the original source than when it was subtracted from the original source. Discrimination did not change as a function of octave band filtering the ripple noise when the center frequency of the filters varied from 200 to 3200 Hz. The results will be discussed in terms of pitch perception, especially in regard to the concept of a spectral dominance region for pitch and pitch strength. The data will also be related to the echo processing abilities of the auditory system. [Work supported by NSF.]
FREE

Detection of signals with identical energy spectra but different pitches under conditions of signal uncertainly (A)

C. Lundeen and A. M. Small, Jr.

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S118-S118 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Previous research has shown that a signal which is one of two pure tones will be less detectable than a signal that is of constant, known frequency throughout. Similar results were obtained in the present study using pulse train signals with identical energy spectra but different pitches. This finding cannot be attributed to the “mistuned” critical‐band mechanism postulated in the single‐band [W. P. Tanner Jr. and R. Z. Norman, IRE Trans. Prof. Group Inf. Theory PGIT‐4, 222–227 (1954)] and multiple‐band [D. M. Green, J. Acoust. Soc. Am. 30, 904–911 (1958)] models of selective auditory attention. Our subjects performed as if they could selectively attend to phase, and/or periodicity pitch cues in order to optimize the detectability of anticipated signals.
FREE

Dichotic perception of virtual pitch: Dependence on subjective lateral position of partials (A)

W. A. Deutsch and R. Steger

J. Acoust. Soc. Am. Volume 60, Issue S1, pp. S118-S118 (1976); (1 page)

Online Publication Date: 11 Aug 2005

Full Text: | Download PDF

Show Abstract
Two general principles of pitch perception, analytic mode resulting in spectral pitch and synthetic mode resulting in virtual pitch have been investigated [E. J. Terhardt, Acoust. Soc. Am. 55, 1061–1069 (1974)]. It is assumed that virtual pitch is a product of Gestalt—perception, associated with some kind of learning process. If this concept is valid, it can be distinguished between spectral‐pitch listeners (type I) and virtual‐pitch listeners (type II). Complex tones consisting of components (a) 220, 330, 440, 550, and 660 Hz presented to the left ear and 550 and 660 Hz to the right ear were followed by (b) 200, 300, 400, 500, and 600 Hz left and 200, 400, and 600 right, which fused partially. The subjects were asked to decide whether pitch (a) or (b) was higher. Type I listeners responded (b) due to the spectral pitch of the corresponding lateral positions of the partials in the head, whereas type II listeners responded (a) referring to the missing fundamental of 110 Hz. In a second run the subjects were asked to pay specific attention to the different lateral positions too and again type I listeners differed from type II. The results are discussed in terms of Gestalt—rules and theory of pitch perception. [Work supported by AKG, Vienna.]
Return to All Sections
Close

Home Publications Meetings Contact ASA Site Map Back to Top

Copyright © Acoustical Society of America

close