Craig C. Wier

Discrimination and labeling of noise–buzz sequences with varying noise‐lead times: An example of categorical perception

The onset of a noise [0.9–2.1 kHz, 55 dB SPL (A weighted)] preceded that of a buzz [100 Hz, 0.5–3.0 kHz, 70 db SPL (A weighted), 500 msec] by −10 to +80 msec and both terminated simultaneously. Eight adults discriminated among noise‐lead times in an oddity task. In separate sessions, they labeled singly presented stimuli with either of the two responses: ’’no noise’’ or ’’noise.’’ The results are highly similar to those reported for the categorical perception of synthetic plosive consonants differing in voice‐onset time. On the average, discrimination was best across a noise‐lead‐time boundary of about 16 msec, where labeling also shifted abruptly. These results and those of categorical perception, generally, are interpreted in terms of Weber’s law as applied to a single component within a stimulus complex. It is concluded that categorical perception of sounds is not unique to speech and suggested that it may be a general property of sensory behavior.Subject Classification: [43]65.75; [43]70.30.

Partial dissociation of spontaneous otoacoustic emissions and distortion products during aspirin use in humans

Otoacoustic emissions (OAEs) of two types--spontaneous and evoked distortion products--were studied before, during, and following a period of aspirin use. As previously reported, aspirin consumption uniformly reduced the spontaneous OAEs (SOAEs) to unmeasurable or extremely low levels. Aspirin consumption also reduced the amplitude of the evoked distortion products (EDPs) but did not eliminate them entirely. The amplitude of the EDP and its change with aspirin consumption were related to both the proximity of the EDP to the frequency of the SOAE and to the level of the primaries producing the EDP. At low primary levels, even with the SOAE absent (due to aspirin consumption, or suppression), EDPs near the SOAE frequency were 10-20 dB higher than when they were 100 Hz away from the SOAE frequency.

Spontaneous narrow‐band oto‐acoustic signals emitted by human ears: A replication

The presence of continuous, narrow‐band acoustic signals (oto‐acoustic emissions) in recordings made from the ear canals of human subjects has been reported by several authors. We report here the results of our attempt to detect such emissions from a series of 92 normally hearing ears from 47 adult subjects. Our results were quite similar to those reported by Zurek [J. Acoust. Soc. Am. 69, 514–523 (1981)]. Emissions were found in 38% of the people and 27% of the ears tested. The distributions of the recorded emissions were also very similar to the results of Zurek’s study.

Detection of a tone burst in continuous‐ and gated‐noise maskers; defects of signal frequency, duration, and masker level

Gated‐noise maskers produce more masking of a tone burst than do continuous‐noise maskers. The difference between the two conditions appears to increase at short signal durations with a maximum difference at about 1000 Hz. The size of the effect is generally small, 1–3 dB, reaching a maximum of 4–7 dB for 1000 Hz at 10 msec and high masker levels.

Discrimination and Labeling of Noise‐Buzz Sequences with Varying Noise‐Lead Times

The onset of a noise (0.9‐2.1 kHz, 55 dB SPL) preceded that of a buzz (100 Hz, 0.5‐3.0 kHz, 70 dB SPL, 500 msec) by —10 to 80 msec and both terminated simultaneously. Eight adults discriminated among noise‐lead times in an oddity task. In separate sessions, they labeled singly presented stimuli with two responses of their choice such as “short noise” and “long noise” or “no noise” and “noise.” The results are highly similar to those reported for the phoneme boundary between voiced and voiceless synthetic plosive consonants. Discrimination was optimal across a 16‐msec noise‐lead time boundary and labeling shifted abruptly at the same noise‐lead time. The results can probably be accounted for in terms of judgment of temporal order or the difference limen for duration of the leading part (noise‐alone) of the stimulus.

Reciprocal relation between the growth of an emitted cubic distortion product and the suppression of a spontaneous otoacoustic emission

Emitted cubic distortion products were generated in ears having a spontaneous otoacoustic emission (SOAE). The frequencies of the primary tones were set to yield a distortion product 20 Hz below the frequency of the SOAE; the levels of the tones were equal and were varied over a wide range. With primaries below about 35 dB SPL, both emissions were present and could be measured separately. As the levels of the primaries were increased, and the amplitude of the distortion product rose, the SOAE was progressively suppressed from its level in the quiet until, at some level of the primaries, it was no longer measurable. Over the range where both emissions were detectable, the rate of decline of the SOAE as a function of the level of the primaries was approximately slope −1.0, while the rate of growth of the emitted distortion product was approximately slope +1.0—suggesting perfect energy conservation. For higher primary levels, where the SOAE was no longer measurable, the growth function showed saturation, being of lower slope, or even nonmonotonic in some cases. Moderate doses of aspirin abolished the SOAEs of all subjects, but only slightly diminished the distortion products. [Work supported by NINCDS Grant NS 15895.]

Gold and Pumphrey revisited, again.

We investigated two periodic sequences: (1) identical sinusoidal pulses repeated regularly in time and (2) a sequence in which alternate pulses were inverted. These are the same stimuli used by Gold and Pumphrey. An analysis of the spectra of these two pulse trains suggested that the critical cue for the discrimination between them is a pitch difference. This pitch difference is the result of a difference in the ripple in the power spectra of the stimuli. This analysis provides an explanation of Gold and Pumphrey’s results that does not require the assumption of extemely narrow tuning, as made in their analysis.Subject Classification: 65.35, 65.60, 65.75.

The relations among critical ratios, critical bands, and intensity difference limens in man

Band-narrowing estimates of the critical bandwidth (CB) are consistently larger than critical-ratio (CR) estimates for the same signal frequency. Bilger [in Hearing and Davis: Essays Honoring Hallowell Davis, edited by S.K. Hirsh et al. (Washington U.P., St. Louis, 1976), p. 191] proposed that this difference could be accounted for by reference to intensity-discrimination performance [CR(Hz)/CB(Hz) = delta I/I]. To test this hypothesis, band-narrowing, critical-ratio, and intensity-discrimination data were collected for four normally hearing, well-trained listeners. Signal frequency was 2000 Hz and two noise levels were used: 20 and 50 dB N0. The relations proposed by Bilger among critical-bandwidth estimates from band-narrowing experiments, critical-ratio estimates from pure-tone detection in wideband noise, and intensity discrimination for a critical-band-wide noise in wideband noise are not supported by the results of individual listeners, or results averaged across listeners.

The relations among critical bands, critical ratios, and intensity difference limens in man

Band‐narrowing estimates of the critical bandwidth (CB) are consistently larger than critical‐ratio (CR) estimates for the same signal frequency. Bilger [in Hearing and Davis: Essays Honoring Hallowell Davis, edited by S. K. Hirsh et al. (Washington U. P., St. Louis, 1976), p. 191] hypothesized that this difference can be accounted for by reference to intensity‐discrimination performance (CF[Hz]/CB[Hz] = ΔI/I). To test this hypothesis, band‐narrowing, critical‐ratio, and intensity‐discrimination data were collected from four subjects. Signal frequency was 2000 Hz and two noise spectrum levels were used: 20 and 50 dB. The psychophysical procedure was an adaptive, two‐interval, forced‐choice, converging on 70.7% correct. Based on critical‐bandwidth estimates obtained from the band‐narrowing experiments using common two‐line regression techniques, critical‐ratio estimates for pure‐tone detection in broadband noise, and intensity‐discrimination for critical bands of noise, the results across experimental cond...

Detection of tones in the absence of external masking noise. II. A reconsideration of the relation between d′ and the ISO standard

Watson, Franks, and Hood [J. Acoust. Soc. Am. 52, 633–643 (1972)] investigated the detectability of tones in the absence of external masking noise in normally hearing listeners and compared their data with the ISO standard for audiometric zero. They concluded that the ISO standard represented thresholds roughly equivalent to d′ = 1.0 performance. Watson and his colleagues used stimuli 150 ms in duration. Pure‐tone temporal‐integration data show that longer signal durations yield lower threshold estimates up to a frequency‐dependent limit. The ISO‐standard data were collected with signals designed to exceed the limits of temporal integration. Psychometric functions for tone detection in quiet are steep; consequently, when longer duration signals are used, a small threshold change of the magnitude expected from temporal integration would alter the performance relation suggested by Watson et al. Re‐evaluation suggests that the ISO standard probably represents a performance level greater than d′ = 2.0. [Work supported by NINCDS.]Watson, Franks, and Hood [J. Acoust. Soc. Am. 52, 633–643 (1972)] investigated the detectability of tones in the absence of external masking noise in normally hearing listeners and compared their data with the ISO standard for audiometric zero. They concluded that the ISO standard represented thresholds roughly equivalent to d′ = 1.0 performance. Watson and his colleagues used stimuli 150 ms in duration. Pure‐tone temporal‐integration data show that longer signal durations yield lower threshold estimates up to a frequency‐dependent limit. The ISO‐standard data were collected with signals designed to exceed the limits of temporal integration. Psychometric functions for tone detection in quiet are steep; consequently, when longer duration signals are used, a small threshold change of the magnitude expected from temporal integration would alter the performance relation suggested by Watson et al. Re‐evaluation suggests that the ISO standard probably represents a performance level greater than d′ = 2.0. [Work ...